TNFR1 signaling promotes pancreatic tumor growth by limiting dendritic cell number and function

Abstract

Pancreatic adenocarcinoma (PDAC) is one the most intractable cancers, in part due to its highly inflammatory microenvironment and paucity of infiltrating dendritic cells (DCs). Here, we find that genetic ablation or antibody blockade of tumor necrosis factor receptor 1 (TNFR1) enhanced intratumor T cell activation and slowed PDAC growth. While anti-PD-1 checkpoint inhibition alone had little effect, it further enhanced intratumor T cell activation in combination with anti-TNFR1. The major cellular alteration in the tumor microenvironment in the absence of TNFR1 signaling was a large increase in DC number and immunostimulatory phenotype. This may reflect a direct effect on DCs, because TNF induced TNFR1-dependent apoptosis of bone-marrow-derived DCs. The therapeutic response to anti-TNFR1 alone was superior to the combination of DC-activating agonistic anti-CD40 and Flt3 ligand (Flt3L). These observations suggest that targeting TNFR1, perhaps in concert with other strategies that promote DC generation and mobilization, may have therapeutic benefits.

Keywords: KPC mice; PDAC; TNF-α; TNFR1; apoptosis; dendritic cells; inflammation; pancreatic adenocarcinoma.

Graphical abstract

Figure 1

Reduced PDAC growth and increased intratumor T cell activation in TNFR1-deficient mice (A–F) KPC cells were subcutaneously implanted in the flank of WT (n = 7), Tnfr1^−/− (n = 9), and Tnfr2^−/− (n = 6) mice. Tumor volumes were measured over time (A) and were harvested for weighing (B) and isolation of infiltrating cells (C). Infiltrating cells were stimulated with ionomycin in the presence of monensin, and IFN-γ and TNF production in CD4⁺ (Thy1.2⁺CD4⁺) and CD8⁺ (Thy1.2⁺CD8β⁺) T cells were determined by intracellular staining and flow cytometry (D). PD-1 expression on CD4⁺ (TCRβ⁺CD4⁺) (E) and CD8⁺ (TCRβ⁺CD8⁺) (F) was measured by flow cytometry. The gating strategy for flow cytometry analyses is shown in Figure S1. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. NS, not significant.

Figure 2

Treatment of subcutaneously injected KPC cells (A–D) KPC cells were subcutaneously implanted in WT or TNFR1-deficient mice, and 2 weeks later, the tumors were injected with control (WT-ctrl, n = 5; Tnfr1^−/−-ctrl, n = 5) or anti-PD-1 antibody (α-PD-1) (WT-α-PD-1, n = 4; Tnfr1^−/−-α-PD-1, n = 5), and tumor progression was monitored. Growth is shown as the percent increase from day 14. On day 38, the tumors were weighed (B) and tumor-infiltrating T cells enumerated (C) and stimulated with PMA/ionomycin to induce cytokine production (D). (E and F) WT mice were injected with KPC cells as in A. After 14 days, the tumors were injected with either control (n = 6), anti-PD-1 (α-PD-1) (n = 6), anti-TNFR1 (α-TNFR1) (n = 6), or both α-PD-1 and α-TNFR1 antibodies (α-PD-1 + α-TNFR1) (n = 5) (E). Growth is shown as the percent increase from day 14. Tumor weight was measured on day 28 (F). ∗p < 0.05, ∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. NS, not significant.

Figure 3

Single-cell sequencing of tumor-infiltrating hematopoietic cells in radiation bone marrow chimeras (A–F) KPC mice were lethally irradiated and reconstituted with either WT (n = 7) or TNFR1-deficient (n = 8) bone marrow. Survival of mice after transfer (day 0) is shown in (A). Single-cell RNA-seq was performed with infiltrating CD45⁺ cells from ∼300 mm³ tumors (B) and DCs (C). Violin plots of the indicated gene expression in DCs (D). GSEA of WT DC2s compared to TNFR1-deficient DC2s. Downward curves represent genes enriched in TNFR1-deficient DCs (E). Uniform manifold approximation and projection (UMAP) plot of CD4⁺ tumor-infiltrating T cells (F). ∗p < 0.05.

Figure 4

Effect of TNFR1 blockade on KPC tumors in RAG2-deficient mice (A) KPC cells were subcutaneously implanted in the flank of WT (n = 5), Tnfr1^−/− (n = 5), and Tnfr2^−/− (n = 5) mice. Tumor-infiltrating DCs were analyzed. (B–E) KPC cells were subcutaneously implanted in RAG2^−/− mice, allowed to grow for 17 days, and then treated with control (n = 5) or anti-TNFR1 antibody (n = 6) every 3 days. Tumor volumes were measured over time (B, left) and tumors removed on day 31 for weighing (B, right) and further analysis (C–E). ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. NS, not significant.

Figure 5

Simultaneous blockade of TNFR1 and PD-1 reduced KPC tumor growth KPC mice in which established pancreatic tumors were detected by ultrasound were treated with either control (n = 7) antibody or anti-TNFR1 plus anti-PD-1 antibody (n = 11). (A–D) Tumors were harvested (ctrl, n = 4; anti-TNFR1, n = 3; and anti-TNFR1 plus anti-PD-1, n = 6) at day 24, and measured tumor volume (A), infiltrating total DCs (B, C), activated DCs (CD11c⁺CD40⁺) (D), and T cells (E) were analyzed. In the figure shown in C, pancreatic tumor tissues were stained for CD11c (green color) and DAPI (blue). Histogram showed the number of CD11c-positive cells per high power field (HPF). (F) Il12b expression of purified DCs measured by quantitative real-time PCR. Each data point represents pooled DCs from 3 mice. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. NS, not significant.

Figure 6

Comparative efficacy of TNFR1 blockade vs. Flt3L and agonistic anti-CD40 in treating subcutaneous KPC cell tumors (A–D) KPC cells were subcutaneously injected into WT or TNFR1-deficient mice and allowed to grow for 17 days. Schematic representation of therapeutic regimen is shown in (A), and the percent tumor growth since day 17 is shown in (B). On day 32, the percentage and number of tumor-infiltrating DCs (C) and their PD-L1 (D) and MHC-II expression (E) were measured. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. NS, not significant.

Figure 7

Effect of TNF on WT and TNFR1-deficient BMDCs (A and B) Bone marrow cells from WT or TNFR1-deficient mice were cultured with GM-CSF in the presence or absence of TNF for 5 days (BMDCs). Cells were counted by light microscopy (A) and PD-L1 expression on CD11c⁺ cells was determined by flow cytometry (B). (C) After 5 days, BMDCs were cultured for another 48 h in the presence of TNF or medium alone and live cells counted (C) and active caspase-3/7 measured by flow cytometry (D). (E and F) After day 5, BMDCs were cultured for an additional 48 h under the indicated conditions live cells (E) and apoptosis measured by staining with annexin V and propidium iodide (PI) (F). (G) DCs were generated from either OT-II or OT-IIxTnfr1^−/− bone marrow as in A, pulsed with OVA whole protein overnight, washed, and 2 × 10⁶ cells intratumorally injected into OT-II KPC tumor-bearing mice. Three days after injection, tumor-infiltrating cells were isolated and analyzed. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. NS, not significant.